# joe-siyuan-qiao/DetectoRS

DetectoRS: Detecting Objects with Recursive Feature Pyramid and Switchable Atrous Convolution

Repository: https://github.com/joe-siyuan-qiao/DetectoRS
Canonical: https://ross.abutalabs.com/products/detectors
Language: Python
License: Apache-2.0
License Family: permissive
Last push: 2021-12-14T04:49:17+00:00

## Health v2 (maintenance only)
Score: 32/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 0, release rhythm 35, longevity 100
- inputs: {"age_days": 2289, "days_push": 1723, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1147, forks 174 (observed 2026-08-28T04:03:45.956311+00:00)

## What it is
Official PyTorch implementation of DetectoRS, a state-of-the-art object detection and instance segmentation model using Recursive Feature Pyramid and Switchable Atrous Convolution, built on the MMDetection codebase. It provides configs and pretrained models achieving strong COCO benchmark results.

## Use cases
- train a state-of-the-art object detection model on COCO
- run instance segmentation with pretrained DetectoRS checkpoints
- apply Recursive Feature Pyramid and Switchable Atrous Convolution to my detector
- reproduce DetectoRS paper results
- integrate RFP and SAC into mmdetection models

## When to choose
- you need top-tier COCO detection/segmentation accuracy with pretrained models
- you already use MMDetection and want to add RFP/SAC components
- you are reproducing or extending the DetectoRS research

## When to avoid
- you need a lightweight or real-time detector
- you want actively maintained code - it is frozen at PyTorch 1.4 and superseded by MMDetection support
- you work outside the MMDetection/PyTorch ecosystem

## Facets
- artifact type: library
- maturity: maintenance
- function: machine-learning, deep-learning, computer-vision, image-processing
- domain: computer-vision, deep-learning, machine-learning
- platform: python
- tags: object-detection, instance-segmentation, mmdetection, pytorch, research-code, coco, linux, gpu

## Member repositories
- joe-siyuan-qiao/DetectoRS (main) score 32

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:03:45.956311+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T06:34:02.820468+00:00, confidence not recorded.
  - readme: https://github.com/joe-siyuan-qiao/DetectoRS (fetched 2026-08-28T04:03:45.956311+00:00, sha e20f80f24174)
- Data as of 2026-08-30T08:39:29.467469+00:00.